View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Servicio de Difusión de la Creación Intelectual

Received: 16 March 2017 | Revised: 26 June 2017 | Accepted: 4 July 2017 DOI: 10.1002/jmor.20735

RESEARCH ARTICLE

The colonic groove of the plains ( maximus): Histochemical evidence of an abrupt change in the glycosylation pattern of goblet cells

María Florencia Tano de la Hoz1,2 | Mirta Alicia Flamini3 | Carolina Natalia Zanuzzi2,3,4 | Alcira Ofelia Díaz1

1Departamento de Biología, Instituto de Investigaciones Marinas y Costeras (IIMyC), Abstract Universidad Nacional de Mar del Plata, The ascending colon of most species shows a longitudinal colonic groove that works as a CONICET, FCEyN, Funes 3250 38 piso, Mar retrograde transport pathway for a mixture of bacteria and mucus toward the cecum. We describe del Plata 7600, Argentina the morphology and glycosylation pattern of the colonic groove of Lagostomus maximus to analyze 2Consejo Nacional de Investigaciones the role of mucins in this anatomical feature. We also studied the distribution pattern of the inter- Científicas y Tecnicas (CONICET), Argentina stitial cells of Cajal (ICC) to evaluate their regulatory influence on gut motility. The groove 3Laboratorio de Histología y Embriología Descriptiva, Experimental y Comparada, originated near the cecocolic junction and extended along the mesenteric side of the ascending Departamento de Ciencias Basicas, Facultad colon, limited at both ends by nonpapillated ridges. These ridges divided the lumen of the ascend- de Ciencias Veterinarias, Universidad ing colon into two compartments: a narrow channel and a large channel, called the groove lumen Nacional de La Plata, La Plata 1900, Argentina and the main lumen, respectively. The histochemical analysis showed differences in the glycosyla- 4Instituto de Patología, “Prof. Dr. Bernardo tion pattern of the goblet cells inside and outside the groove. Unlike the mucosa lining the main Epstein”, Facultad de Ciencias Veterinarias, lumen of the colon, the groove was rich in goblet cells that secrete sulfomucins. The PA/Bh/KOH/ Universidad Nacional de La Plata, La Plata PAS technique evidenced an abrupt change in the histochemical profile of goblet cells, which 1900, Argentina presented a negative reaction in the groove and a strongly positive one in the rest of the colonic

Correspondence mucosa. The anti-c-kit immunohistochemical analysis showed different ICC subpopulations in the Alcira Ofelia Díaz, Instituto de ascending colon of L. maximus. Of all types identified, the ICC-SM were the only cells located Investigaciones Marinas y Costeras (IIMyC), solely within the colonic groove. FCEyN, CONICET-Universidad Nacional de Mar del Plata. Funes 3250 38 piso, 7600 Mar del Plata, Buenos Aires, Argentina. KEYWORDS Email: [email protected] colonic separation mechanism, glycoconjugates, , interstitial cells of Cajal, morphology Funding information This research was supported by a Grant from Universidad Nacional de Mar del Plata (EXA 765/16), Buenos Aires, Argentina.

1 | INTRODUCTION to the existence of a colonic separation mechanism (CSM). Even though the identification of both types is evident in lagomorphs Cecotrophy is a physiological mechanism described in small herbivore (rabbits and hares) the formation of cecotrophs has also been species and it consists of the production and ingestion of a special type demonstrated in caviomorph (Martino, Zenuto, & Busch, 2007; of feces derived from caecal contents (Bjornhag€ & Snipes, 1999; Mess & Ade, 2005; Takahashi & Sakaguchi, 1998, 2000). Sakaguchi, 2003). Diverse studies have demonstrated that ingested Unlike lagomorphs, which possess a “particle dependent” CSM, feces have a larger content of proteins, vitamins and bacteria as well as most rodents have “mucus dependent” CSM; that is, bacteria are a lesser proportion of fibers than the nonswallowed feces (Takahashi & trapped by mucus and transferred through the colonic groove by anti- Sakaguchi, 1998, 2000). The formation of two types of feces, one with peristalic movements into the cecum (Kotze, van der Merwe, Ndou, high protein content (soft feces or cecotrophs) and another fiber-rich O’Riain, & Bennett, 2009; Takahashi & Sakaguchi, 2000). As a result of (hard feces), is possible in different species of herbivore due the CSM, the bacteria do not release with feces but they accumulate in

Journal of Morphology. 2017;1–13. wileyonlinelibrary.com/journal/jmor VC 2017 Wiley Periodicals, Inc. | 1 2 | TANO DE LA HOZ ET AL. the cecum, the site of bacterial fermentation. The material separated article, we aim to study the morphological and histochemical character- by CSM is excreted because of the cecum contraction, and then istics of the colonic groove of L. maximus to analyze the role of mucins reingested. in this anatomical feature. In addition, we described the pattern of dis- In general, the main method to demonstrate cecotrophy is the tribution of the interstitial cells of Cajal (ICC), that is, cellular pace- direct observation of certain positions and movements characteristic of makers in close contact with nerve cells in the muscular layer, to this behavior (the majority of rodents bend the head to bring the determine a possible implication between their distribution and the mouth to the anus), as well as the differential analysis of soft and hard regulation of the motility of the colonic groove. These cells are a critical feces. However, this behavior is difficult to observe in some wild component in enteric neuromuscular transmission. Advances in the last rodents with fossorial and nocturnal habits such as in the case of the decades have shown that ICC play an important role in coordinating (Lagostomus maximus) (Clauss, Besselmann, Schwarm, intestinal motility (Mazzone & Farrugia, 2007). Ortmann, & Hatt, 2007). The plains viscacha (Desmarest, 1817) is a caviomorph rodent belonging to the Chinchilidae that lives in arid, | semiarid and humid regions of Argentina, Bolivia, and Paraguay. It is a 2 MATERIALS AND METHODS gregarious that presents fossorial habits and lives in communal 2.1 | cave systems known as vizcacheras (Jackson, Branch, & Villarreal, 1996). At twilight they leave their caves in search of food, especially Adult , Lagostomus maximus (Desmarest, 1817) of both sexes grasses and dicotyledons (Bontti, Boo, Lindstrom,€ & Elia, 1999; Branch, (n 5 14, 8 females and 6 males) weighing between 4 and 5.5 kg were Villarreal, Sbriller, & Sosa, 1994; Puig, Videla, Cona, Monge, & Roig, obtained from the Estacion de Cría de Animales Silvestres (ECAS; Wild 1998). Recent studies using markers for digestibility assays have shown Animals Breeding Station), Ministry of Agro Industry of the Province of that L. maximus, like other hystricomorph rodents has a longitudinal Buenos Aires (Argentina). The captured animals were anesthetized with colonic groove and re-ingests its own feces (Clauss et al., 2007; Hagen a dose of xylacine (8 mg/kg body weight) followed by ketamine et al., 2015). Also, studies based on passage of ingest have shown the (50 mg/kg body weight) by intramuscular injection (Ketanest, Labora- existence of “mucus dependent” CSM in L. maximus (Hagen et al., torio Scott Cassara). Once deep anesthesia was reached, intracardiac 2015). The available information on the digestive physiology of perfusion with physiological saline solution and then with 4% parafor- L. maximus makes it an interesting species for the study of this maldehyde in 0.1 mol L-1 phosphate buffer was performed. The proto- morphological adaptation to herbivory. col was approved by the Institutional Committee for the Care and Use The mucus is a permeable gel layer, mainly composed of water and of Laboratory Animals at the National University of La Plata (52–4- mucins; it covers the mucosa of the vertebrate gastrointestinal tract. 15T) and was in compliance with the international recommendations Mucins are highly glycosylated proteins Bansil and Turner, 2006 syn- for experimental animals (Commission on Life Sciences National thesized chiefly by goblet cells of the intestinal epithelium. It was dem- Research Council, 1996; Zuniga,~ Tur Marí, Milocco, & Pineiro, 2001). onstrated that among their functions they enable nutrient interchange with the underlying epithelium, protect the mucosa against proteolytic | injury and are the main attachment site of commensal and pathogenic 2.2 Sampling and morphological study bacteria (Hansson, 2012; Kim & Ho, 2010). The necropsy was performed immediately after euthanasia by isolating In vertebrates, mucins exhibit different glycosylation patterns due to the ascending colon. The inner mucosa was exposed with a cut along the diverse length, branching and acetylation of the oligosaccharide the antimesenterial border. Once the colonic mucosa was exposed, ’ chains (Forstner, Oliver, & Sylvester, 1995). According to the glycans photographs were taken of the colonic groove along its entire length. chemical characteristics, mucins can be classified into neutral and acidic. The shape of the grooves was recorded. Transverse sections of the cra- In turn, acidic mucins can be subdivided into sulfomucins or sulfated nial portion, the medial third, and the terminal portion of the ascending mucins, and sialomucins (glycoproteins with sialic acid residues) (Beyaz & colon were taken for histological and histochemical analysis. Samples Liman, 2009). The physicochemical characteristics of mucins are deter- were routinely processed and embedded in paraffin wax. Histological mined mainly by their composition and their content of glycans (Liquori sections of 4 mm thickness were stained with hematoxylin-eosin (H-E). et al., 2012). Consequently, the analysis of the glycosylation pattern of Microphotographs were taken with an Olympus microscope, CH30 mucins has been widely used in diverse studies to infer its physiological (Olympus; www.olympus.com). role (Mastrodonato, Mentino, Liquori, & Ferri, 2013; Scillitani & Mentino, 2015; Tano de la Hoz, Flamini, & Díaz, 2014, 2016). Although it has been suggested that mucus play a key role in bac- 2.3 | Histochemistry teria transport through the longitudinal colonic groove, there is paucity For the characterization of glycoconjugates (GCs) the histological sec- in the literature of studies determining the histochemical characteristics tions were also subjected to the following histochemical techniques: of mucins in that region. Moreover, no exhaustive research on the rodents’ histochemical profile of the colonic groove has been reported 1. Periodic acid Schiff (PAS) to demonstrate GCs with oxidizable vici- (Kotze et al., 2009; Snipes, Hornicke,€ Bjornhag,€ & Stahl, 1988). In this nal diols and glycogen (Mc Manus, 1948). TANO DE LA HOZ ET AL. | 3

2. a-amylase-PAS. Before the PAS technique, sections were sub- 2.4 | Lectin-histochemistry jected to an enzymatic digestion with a-amylase for glycogen A battery of seven biotinylated lectins (Vector Laboratories, Inc. identification (Pearse, 1985). Burlingame, CA) was used to identify specific sugar residues (Table 3. KOH/PA*S (saponification- selective periodic acid- Schiff’s 1). The paraffin sections were mounted on slides treated with poly- reactive) for characterization of GCs with sialic acid residues. The L-lysine (Sigma Diagnostics, St Louis, MO), deparaffinated with xylol saponification reaction (KOH) was performed with 0.5% sodium and incubated in a 0.3% H2O2 solution in methanol for 30 min at hydroxide in 70% ethanol for 30’ at room temperature. Before the room temperature to inhibit the endogenous peroxidase activity. Schiff’s reactive, sections were subjected to a selective oxidation 2 Then, the sections were hydrated, washed with 0.01 mol L 1,pH with 0.4 mmol L-1 periodic acid in 1.0 mmol L-1 hydrochloric acid 7.6 saline phosphate buffer (SPB) and incubated with a bovine at 48C (Culling, Reid, & Dunn, 1976). serum albumin in SPB for 20 min to block specific bindings. After 4. PA/Bh/KOH/PAS (periodic acid - reduction with borohydride - rinsing, sections were incubated with each of the biotinylated lectins saponification- periodic acid- Schiff’s reactive) for presence of for 30 min at room temperature and treated with avidine–biotine– 7 8 GCswithsialicacidresidueswithO-acylsubstitutionsin C, C peroxidase complex (ABC) during 45 min (Vector Laboratories, Inc). 9 o C, and O-acyl sugars. This method was carried out using a 2 The peroxidase was activated through 4–10minincubationwitha hr oxidation at room temperature with 1% periodic acid. The Tris-HCl 0.05 mol L21, pH 7.6 tamponed solution containing 0.02% aldehydes generated by the initial oxidation were reduced to diaminobenzine (DAB; Dako, Carpinteria, CA, EE.UU.), and 0.05% H2O2. primary alcohols with sodium borohydride. After saponification All lectins were employed in a 30 mg ml-1 in SPB, except for PNA that (KOH) the PAS technique (Reid, Culling, & Dunn, 1973) was was applied at concentration 10 mg ml-1. applied. The results were semi-quantitatively evaluated, using the same 5. KOH/PA*/Bh/PAS (saponification- selective periodic acid- reduc- scale as the one used for the histochemical study. Two types of con- tion with borohydride- periodic acid- Schiff’s reactive) to identify trols were made: (1) the lectin solution was replaced by SPB and (2) lec- neutral sugars. Sections were treated with 0.5% potassium tins were preincubated at ambient temperature for 1 hr in presence of hydroxide in 70% ethanol for 15 min at ambient temperature. adequate haptens. Previous to the PAS technique a selective periodic oxidation at 48C for 1 hr followed by a reduction with sodium borohydride was 2.5 | Immunohistochemical detection of performed (Volz, Reid, Park, Owen, & Dunn, 1987). interstitial cells of Cajal 6. AB (Alcian Blue) is a basic dye that has affinity for tissue compo- For the immunohistochemical analysis of immunohistochemical detec- nents that possess anionic groups such as acidic GCs. AB solutions tion of interstitial cells of Cajal we used the Envision (Dako Corp.,Car- at different pH were used to selectively stain subgroups of acidic pinteria, CA) method. A rabbit polyclonal antibody anti-c-kit (CD117; mucins. A staining solution (pH 2.8) was used to evidence GCs with Dako, Japan A4502) was used as a primary antibody to detect ICC. The carboxylic groups and O-sulfated esters; sulfated GCs were identi- paraffin sections were mounted on Starfrost glass slides (Knittel, fied with a pH 1.0 solution and GCs highly sulfated with a pH 0.5 Braunschweig, Germany), deparaffinated with xylol and incubated in (Lev & Spicer, 1964). 0.3% H2O2 in methanol for 30 min at room temperature. Then they 7. AB/PAS (Alcian Blue - periodic acid Schiff). This combined tech- were hydrated, washed with 0.01 mol L-1, pH 7.6 SPB, and microwave nique allows the identification of acid (AB positive), neutral (PAS- antigen retrieval was applied at 800w twice for 5 min each using positive), and mixed (AB/PAS positive) GCs in one section Mowry, 0.01 mol L-1, pH 6.0 buffer citrate. After rinsing, sections were incu- 1963. The AB solution was used at pHs 2.8 and 1.0 to identify bated with 1% bovine seric albumin for 30 min to block unspecific carboxylated and sulfated GCs, and GCs with O-sulfated esters, bindings. The sections were incubated with the primary antibody (1:50 respectively (Mowry, 1963). dilution) for 1 hr at 258C; negative controls were incubated with SPB 8. TB (toluidine blue). This basic dye has affinity for acidic compo- under the same conditions. After incubation with the polymer, 3’,3’-dia-

nents of tissues, such as chromatin. Glycoconjugates with carbox- minobenzidine (DAB) and 0.05% H2O2 were used. Finally, sections ylic groups and O-sulfated esters were evidenced with an TB were counterstained with hematoxylin (BIOPUR), dehydrated and solution at pH 5.6 and sulfated GCs at pH 4.2 (orthochromatia). mounted. Moreover, this dye can stain polyanionic polymers of high molecu- lar weight with a color different from the original tint (metachro- 3 | RESULTS matia) (Lison, 1953). 3.1 | Morphological study The results were evaluated by four independent observers using a semi-quantitative scale to determinate the intensity of the reactions The ascending colon of Lagostomus maximus exhibited a longitudinal (0, negative; 1, light; 2, moderate; 3, strong). These scores were colonic groove along the mesenteric side. The groove originated near established according to previous histochemical studies (Scillitani & the cecocolic junction and it was formed by two nonpapillated ridges Mentino, 2015; Tano de la Hoz et al., 2014, 2016). (Figure 1a,b). These ridges divided the lumen of the ascending colon 4 | TANO DE LA HOZ ET AL.

TABLE 1 Lectins used and their carbohydrate specificities

Lectin Acronym Specificity Haptene

GROUP I Glc/Man

Canavalia ensiformisagglutinin Con-A a-D-Man; a-D-Glc a-D-Methyl-Man

GROUP II GlcNAc

Triticum vulgaris WGA b-D-GlcNAc; NeuNAc NeuNAc (wheatgerm) agglutinin

GROUP III GalNAc/Gal

Dolichos biflorusagglutinin DBA a-D-GalNAc D-GalNAc

Glycine maximusagglutinin SBA a-D-GalNAc; b-D-GalNAc D-GalNAc

Ricinus communis RCA-I b-Gal Lactose Agglutinin-I

Arachis hypogaeaagglutinin PNA b-D-Gal (b1–3) > D-GalNAc Lactose

GROUP IV L-Fuc

Ulex europaeus UEA-I a-L-Fuc L-Fuc Agglutinin-I

Gal, galactose; GalNAc, N-acetylgalactosamine; Glc, glucose; GlcNAc, N-acetylglucosamine; L-Fuc, L-fucose; Man, mannose; a-D-Methyl-Man, a-D-Methyl-mannose; NeuNAc, acetyl-neuraminic acid (sialic acid).

FIGURE 1 Lagostomus maximus, histological characterization of the longitudinal colonic groove. (a) Cross section of the ascending colon. (b) Macroscopic anatomy of the groove’s origin. (c) Macroscopic view of the inner surface of the ascending colon. (d) Macroscopic view of the caudal region of the ascending colon. Arrow head, caudal region of the ascending colon where the groove disappeared; asterisk, cecocolic junction; Ce, cecum; G, groove; MLu, main lumen; R, ridge. Scale bar: 1 cm (a); 1.5 cm (b, c, d) TANO DE LA HOZ ET AL. | 5

FIGURE 2 Lagostomus maximus, histological characterization of the longitudinal colonic groove, H-E. (a) Microphotography of the groove. (b) Nonpapillated ridge. (c) Detail of the abrupt change of the tunica muscularis morphology. (d) Detail of the tunica muscularis. Arrow head, large caliber vein; block arrow, morphological change of the tunica muscularis; CLS, lateral border of the ridge limiting with the main lumen; G, groove; GS, lateral border of the ridge limiting the groove; M, tunica mucosa; MLu, main lumen; R, ridge; SubM, tunica submucosa; TM, tunica muscularis; TMC, inner circular layer of the tunica muscularis; TML, outer longitudinal layer of the tunica muscularis. Scale bar: 800 mm (a); 400 mm (b); 250 mm (c); 100 mm(d) into two compartments: a narrow channel and a large channel, called 3.2 | Histochemical study the groove lumen and the main lumen, respectively (Figure 1a). The The analysis of glycoconjugates (GCs) revealed evident differences space separating the ridges gradually diminished along the colon between the glycosylation patterns of goblet cells in and out of the producing a narrowing of the groove toward the distal region of the colonic groove all along the ascending colon (Table 2). The change in ascending colon, where it finally disappeared (Figure 1c,d). Every histochemical profile of this cell type occurred with most techniques at ridge showed two lateral edges: one that lines the main lumen and the height of the colonic ridges. In all the specimens studied the another that limits with the colonic groove named colonic luminal mucosa lining the colonic luminal side of the ridge (CLS) showed a gly- side (CLS) and groove side (GS), respectively (Figures 2a and 3). The cosylation pattern different to the one lateral to the ridge limiting the center of the ridge consisted of an extension of the tunica submu- groove (GS; Figure 4). cosa characterized by large caliber veins (Figure 2b). The morphol- The AB pH 2.8/PAS method allowed the identification of goblet ogy of the tunica muscularis of the ascending colon abruptly cells containing neutral, acidic and mixed mucins, both in the groove and changed at the height of the ridges. Instead of being formed by an the main lumen of the organ (Figure 4a–d). However, the distribution inner circular layer and an outer longitudinal layer, as in most regions of the digestive tract, the tunica muscularis of the groove exhibited pattern of these unicellular glands differed noticeablybetweenbothsec- numerous bundles of circularly arranged smooth muscular fibers tors (Figure 4c,d). Goblet cells out of the groove showed a decreasing separated by abundant connective tissue (Figure 2a,c). Instead, the gradient of neutral GCs from the upper to the lower region of the crypts, longitudinal outer layer gradually diminished its width until it finally and PAS and AB/PAS positive cells were found in the upper third of the disappeared within the groove (Figure 2d). Although the tunica mus- gland. Goblet cells from the basal zone showed a histochemical pattern cularis of the groove presented only an inner circular layer, it was different to the rest of the cells, exhibiting mainly carboxylated GCs (Fig- characterized by being thicker than the outer muscular layer of the ure 4c). In the groove, instead, all three cell types were identified, ascending colon. although distributed along all the axis of the intestinal gland (Figure 4d). 6 | TANO DE LA HOZ ET AL.

FIGURE 3 Scheme of general histological characteristics of the colonic groove. Asterisk, folds; CSL, lateral border of the ridge limiting the main lumen; CT, connective tissue; M, tunica mucosa; MLu, main lumen; G, groove; GS, lateral border of the ridge limiting the groove;

SubM, tunica submucosa; TMC, inner circular layer of the tunica muscularis; TML, outer longitudinal layer of the tunica muscularis; v, large caliber vein. Scale bar: 2000 mm

Unlike the mucosa which lines the main lumen of the colon, the Goblet cells from both regions showed a moderate staining with groove was characterized by secreting a high proportion of sulfomucins the KOH/PA*S technique evidencing GCs with sialic acid residues (Fig- and GCs with polyanionic complexes (Figures 4e,f and 5a). Even though ure 5c–e). However, the PA/Bh/KOH/PAS method showed an abrupt goblet cells from the main lumen also showed metachromatia with TB, change in the histochemical profile of the cells, which presented a neg- at pHs both 5.6 and 4.2, this reaction was only observed in cells from ative reaction in the groove and a strongly positive reaction in the rest the upper and medium region of the intestinal crypts (Figure 5b). of the colonic mucosa. These results demonstrated that only goblet

TABLE 2 Histochemical analysis of the ascending colon of Lagostomus maximus

Ascending colon Groove Main lumen Goblet cells Goblet cells

Techniques Glycocalyx Uppercrypt Medialcrypt Lowercrypt Glycocalyx Upper crypt Medialcrypt Lowercrypt

PAS03 3 3 03 3 1

AB pH 2,8 0 3 3 3 0 3 3 3

AB pH 1,0 2 3 3 3 1 1 1 1

AB pH 0,5 2 3 3 3 1 1 1 1

AB pH 2,8/PAS 0 3M-3P-3Ba 3M-3P-3Ba 3M-3P-3Ba 0 3M-3Pa 3P 3B

AB pH 1,0/PAS 1 3M-3P-3Ba 3M-3P-3Ba 3M-3P-3Ba 1 2P-2Ma 2P 1P

TB pH 5,6 0 3m 3m 3m 0 3m 3m 1or

TB pH 4,2 0 3m 3m 3m 0 3m 3m 1or

KOH/PAaS0 2 2 2 0 2 2 2

PA/Bh/KOH/PAS 0 0 0 0 0 2 2 2

KOH/PAa/Bh/PAS 0 3 3 3 0 3 3 3 m, metachromasia; M, magenta; or, ortochromasia; P, purple. Staining intensity: 0, negative; 1, slightly positive; 2, moderate; 3, strong. aGoblet cells with different histochemical profiles were differentiated. TANO DE LA HOZ ET AL. | 7

FIGURE 4 Lagostomus maximus, in situ characterization of the glycosylation pattern of the ascending colon. (a) Microphotography of a colonic ridge, AB pH 2.8/PAS. (b) Detail of figure (a) showing the abrupt change in the histochemical profile of goblet cells (see dotted line), AB pH 2.8/ PAS. (c) Microphotography of the lateral ridge limiting the main lumen of the colon (CLS), AB pH 2.8/PAS. (d) Microphotography of a intestinal crypt in the side of the ridge limiting the colonic groove (GS), AB pH 2.8/PAS. (e) Nonpapillated ridge showing an abrupt change in the histochemical pattern of goblet cells (see dotted line), AB pH 1.0/PAS. (f) Detail of figure e, AB pH 1.0/PAS. Arrow head, large caliber vein; CLS, lateral border of the ridge limiting the main lumen; G, groove; GoA, goblet cell AB/PAS positive; GoB, goblet cell PAS positive; GoC,gobletcellAB positive; GS, lateral border of the ridge limiting the groove; MLu, main lumen; R, ridge. Scale bar: 500 mm(a,e);250mm (b, f); 75 mm(c);50mm(d) cells from the main lumen secrete GCs with sialic acid O-acyl substi- intensively labeling the glycocalyx and the goblet cells of the upper and tuted at C7, C8, or C9 (Figure 5f,g). medium regions of the intestinal glands (Figure 6e,f). Although the SBA lectin strongly labeled the glycocalyx of both sectors, it only showed 3.3 | Lectin histochemical study affinity for goblet cells outside the groove (Figure 7a,b). Conversely, UEA-I evidenced L-fucose residues in goblet cells of the whole organ, The lectin histochemical method revealed different specific sugar resi- except in cells in the base of the intestinal glands of the main lumen dues in the groove and the main lumen of the ascending colon (Table (Figure 7g,h). 3). The Con-A, WGA, RCA-I, and PNA lectins showed the same binding pattern in both sectors, exhibiting an intense labeling both in the glyco- 3.4 | Immune histochemical study to detect calyx and the goblet cells (Figures 6a–dand7c–f). However, the colo- interstitial cells of Cajal nic groove showed a lectin histochemical profile that differs from the rest of the ascending colon in all the studied regions. The DBA lectin The ascending colon wall lining the main lumen presented cells strongly gave a negative reaction in the groove but a positive in the main lumen immunolabeled in the myenteric plexus and throughout the thickness 8 | TANO DE LA HOZ ET AL.

TABLE 3 Lectin histochemical analysis of the ascending colon of Lagostomus maximus

Ascending colon

Lectin Groove Main lumen Glycocalyx Goblet cells Glycocalyx Goblet cells

Con-A 2 2 2 2

WGA3232

DBA0033a

SBA3033

RCA-I 3 2 3 2

PNA3232

UEA-I 1 2 1 2a

Staining intensity: 0, negative; 1, slightly positive; 2, moderate; 3, strong. aOnly the cells of the upper and middle region of the intestinal crypts presented positive reaction.

4 | DISCUSSION

4.1 | Comparison of the colonic groove morphology within hystricomorph rodents

The colonic separation mechanism (CSM) has been studied in differ- ent species of lagomorphs and rodents (Bjornhag€ & Snipes, 1999; Hagen et al., 2015; Takahashi & Sakaguchi, 2000). However, the his- FIGURE 5 Lagostomus maximus, in situ characterization of the glycosylation pattern of the ascending colon. (a) Mucosa and tological characteristics of this adaptation to herbivory have been submucosa of a colonic ridge, TB pH 5.6. (a) Detail of the goblet described only in few studies (Kotze et al., 2009; Snipes et al., 1988). cells at the basal region of the intestinal glands, TB pH 5.6. (b) Our results show that although the anatomical and histological orga- Mucosa lining the main lumen of the colon, TB pH 5.6. (c) Ridge of nization of the colonic groove of Lagostomus maximus is similar to the colonic groove, KOH/PA*S. (d) Mucosa lining the main lumen, that of other hystricomorph rodents, this species also possesses its KOH/PA*S. (e) Mucosa lining the colonic groove, KOH/PA*S. (f) own distinctive features. The colonic ridges of L. maximus has a cen- Microphotography of a colonic ridge showing an abrupt change in the histochemical pattern of goblet cells (see dotted lines), PA/Bh/ ter of connective tissue with irregular large caliber veins (Kotzeetal., KOH/PAS. (g) Detail of figure f. arrow head, large caliber vein; 2009; Snipes et al., 1988). It is possible, as proposed by Kotzeetal. black arrow, goblet cells with orthochromatic reaction; CSL, lateral (2009) as seen in all rodent species that have been studied. These border of the ridge limiting the main lumen; Go, goblet cells; GS, large veins supposedly generate the swelling of the ridges as a con- lateral border of the ridge limiting with the groove; R, ridge; white sequence of the high blood irrigation, and thus compartmentalize the arrow, goblet cells with metachromatic reaction. Scale bar: 100 mm (a); 50 mm(b);40mm (d, e); 250 mm (c, f, g) lumen. Because the existence of these large veins has been described in every studied species, the groove closure mechanism of the inner circular layer and the outer longitudinal layer of the tunica may be common to all rodents. In some species of the African mole muscularis (Figure 8). Because of their location and morphology these rats (Rodentia, Bathyergidae) the presence of projections at the end cells are classified as myenteric ICC (ICC-MY) and intramuscular ICC of the ridges that would further facilitate the closure of the groove (ICC-IM), respectively. The ICC-MY showed a multipolar morphology (Kotze et al., 2009). In contrast, no additional structures have been (Figure 8b) whereas the ICC-IM presented a fusiform body with few found in the ridges of L. maximus and Myocastor coypus to contribute branched processes orientated according to the direction of the major to the separation of the main lumen of the groove. axis of the muscle fiber (Figure 8a,c). Differences among species have also been observed in the struc- In the colonic groove ICC-IM were identified only in the inner cir- tural organization of the tunica muscularis. Contrary to the description cular layer (this region of the ascending colon lacks a longitudinal layer). made by Kotze et al. (2009) on the six species of Bathyergidae, in our Moreover, numerous c-kit positive cells were detected in the submuco- study the tunica muscularis of L. maximus showed significant variations sal surface of the circular muscle (ICC-SM; Figure 9). This ICC- in both muscle layers in the colonic groove. It showed an inner circular subpopulation extended from ridge to ridge, forming a continuous layer formed by bundles of smooth muscle fibers separated by abun- band of multipolar cells with thin, long, and interdigitated processes dant connective tissue whereas the longitudinal outer layer gradually (Figure 10). decreased in thickness and disappeared at the height of the ridges. TANO DE LA HOZ ET AL. | 9

L. maximus may act as an individual functional unit that works together to carry out specific functions.

4.2 | Glycosylation pattern of L. maximus colonic groove: Functional implications

Several studies have demonstrated that rodents transport through the CSM a mixture of bacteria and mucus from the ascending colon to the cecum to maintain the bacterial fermentation process (Snipes et al.,

FIGURE 6 Lagostomus maximus, lectin histochemistry of the longitudinal colonic groove (a, c, e) and of the ascending colon wall lining the main lumen (b, d, f). (a, b) Con- A. (c, d) WGA. (e, f) DBA. Arrow, glycocalyx; Go, goblet cell; mm, muscular mucosa. Scale bar: 20 mm(a);30mm(b);60mm (c); 40 mm(d–f)

Similar histological characteristics were described for M. coypus although the organization of the bundles of the inner circular layer dif- fered from that of L. maximus (Snipes et al., 1988). In contrast, an increase in thickness of the outer longitudinal layer exactly beneath the ridges that limit the groove was observed in six species of naked mole- rats (Kotze et al., 2009). Although noticeable differences among species were observed, rodents show changes in the histological characteristics of the tunica muscularis at the height of the ridges. The high structural FIGURE 7 Lagostomus maximus, lectin histochemistry of the specialization of the muscle tissue is probably linked to the antiperistal- longitudinal colonic groove (a, c, e, g) and of the ascending colon tic movements in the colonic groove that transport the luminal content wall lining the main lumen (b, d, f, h). (a, b) SBA. (c, d) RCA-I. (e, f) in a retrograde way (Takahashi & Sakaguchi, 2006). It is likely that each PNA. (g, h) UEA-I. Arrow, glycocalyx; Go, goblet cell. Scale bar: 30 muscle bundles of the inner circular layer of the tunica muscularis of mm(a–c); 25 mm(d–g); 50 mm(h) 10 | TANO DE LA HOZ ET AL.

by the groove goblet cells, no exhaustive study on the histochemical characteristics has been done until now. This is the first complete anal- ysis on the glycosylation pattern of mucins secreted in and out of the colonic groove, since so far, most research has focused mainly on the morphological and physiological characteristics of this anatomic adap- tation (Kotze, van der Merwe, & O’Riain, 2006; Kotze et al., 2009; Snipes et al., 1988; Takahashi & Sakaguchi, 2000, 2006). The GCs analysis showed differences among goblet cells in and out of the colonic groove, this being the first time an abrupt change in the glycosylation pattern of the intestinal tract of L. maximus is described (Tano de la Hoz et al., 2014, 2016). The observed varia- tion of the ascending colon histochemical profile at the ridges level suggests that the glycosylation pattern of mucus plays a key role in the functioning of the groove of L. maximus. As described by Hans- son (2012), mucus can trap bacteria in the intestinal tract lumen in many ways. First, bacteria can get trapped into the mucin polymeric network or they may bind to the great variety of mucus glycans through fimbrial and afimbrial adhesins which specifically recognize different types of carbohydrate residues. Because of the signifi- cance of mucus composition in bacterial aggregation, it is possible that the distinct histochemical profile of the colonic groove gener- ate specific anchorage sites for the high density of bacteria in the region. In addition, as it has been demonstrated in other research, the presence of highly glycosylated mucins may limit the access of pathogenic bacteria to the cell surface by steric hindrance (McGuckin, Linden, Sutton, & Florin, 2011). Published evidence shows that mucins interact with the intesti- nal microflora dynamically and adaptively (Corfield, 2015). Freitas, Axelsson, Cayuela, Midtvedt, & Trugnan (2002) have shown that the microbiota induces modifications of the intestinal mucin glycosyla- tion to produce an increase secretion of sulfated mucins. Similarly, our results documented that the mucus secreted by goblet cells of the colonic groove displays a greater proportion of sulfomucins than that produced by the rest of the goblet cells from the ascending colon. Diverse studies have demonstrated that sulfate groups confer mucus a greater resistance to degradation by bacterial glycosidases and host proteases (McGuckin et al., 2011; Roberton & Wright, 1997). Moreover, studies on the ontogenetic development of the mammalian intestinal tract have demonstrated an increase in acidic mucin secretion during the fetal stages that would probably contrib- FIGURE 8 Lagostomus maximus, immunohistochemistry anti-c-kit ute to improve the innate immunological response in the prenatal of the ascending colon wall lining the main lumen. (a) Positive cells in the circular layer of the tunica muscularis. (b) Positive cells at stages (Beyaz & Liman, 2009). In view of the foregoing background the myenteric plexus level. (c) Positive cells in the longitudinal and because the colonic groove lumen transports more bacteria than layer of the tunica muscularis. Arrow, cytoplasmic prolongation; the main lumen of the ascending colon (Takahashi & Sakaguchi, arrow head, ICC; TMC, inner circular layer of the tunica muscularis; 2006), it is possible that the sulfomucins secreted in the L. maximus TM , outer longitudinal layer of the tunica muscularis. Scale bar: L colonic groove may be implicated in the protection of mucosa to pre- 30 mm vent the proliferation of pathogenic bacteria. As for the different types of sialomucins secreted by goblet cells, 1988; Takahashi & Sakaguchi, 2000). Because the selective transport the histochemical analysis also revealed variations between the colo- of bacteria to the cecum is produced by mucins secreted in the colonic nic groove and the main lumen. Although the presence of GCs with groove the mechanism is considered mucus dependent (Sakaguchi, sialic acid residues was evidenced in all goblet cells from the ascend- 2003). Although the CSM largely depends on the mucus synthesized ing colon, our results demonstrated that only in the main lumen of L. TANO DE LA HOZ ET AL. | 11

FIGURE 9 Lagostomus maximus, immunohistochemistry anti-c-kit of the colonic groove. (a) Microphotography of a ridge showing a contin- uous band of c-kit positive cells in the submucosa region limiting the tunica muscularis. (b, c) Detail of the tunica submucosa with c-kit pos- itive cells. Arrow, ICC-SM; arrow head, large caliber vein; CLS, lateral border of the ridge limiting the main lumen; G, groove; Gs, lateral border of the ridge limiting the groove; MLu, main lumen; mm, muscular mucosa; R, ridge; SubM, tunica submucosa; TM, tunica muscularis. Scale bar: 500 mm (a); 100 mm(b,c)

maximus GCs with sialic acid O-acyl substituted at C7, C8, or C9 are enteric neurons, the interstitial cells of Cajal (ICC) and the smooth mus- secreted. Similarly, other studies have described a high proportion of cle cells (Mazet, 2015). Advances in the last decades have improved O-acetylated sialomucins in the colon of humans and rodents the comprehension of the ICC role in the GIT, and shown that they are (Accili, Menghi, & Gabrielli, 2008; Mastrodonato et al., 2013). As an integral part of the gastrointestinal motor apparatus (Mazzone & previously proposed, the acetylation of sialic acid residues can Farrugia, 2007). However, to the present there are no studies on the substantially modify its functional role in the diverse biological proc- ICC-distribution pattern in the colonic groove of rodents. According to € esses (Angata & Varki, 2002). The higher degree of acetylation of the classification made by Sanders, Ordog,€ Koh, Torihashi, & Ward sialic acid residues secreted by goblet cells from the main lumen alter (1999), the immunohistochemical detection of ICC demonstrated the the viscoelastic and protective properties of the mucus, and thus existence of different subpopulations in the ascending colon of L. maxi- improving the resistance to bacterial neuraminidases (Mastrodonato mus. Considering the numerous physiological studies performed on the et al., 2013). ICC functional role (Mazzone and Farrugia, 2007; Sanders, Kito, The lectin binding pattern also presented some variations between Hwang, & Ward, 2016; Sanders et al., 1999), we can infer that the the colonic groove and the main lumen of the ascending colon of L. diverse subtypes found would participate in the intestinal motility of L. maximus, in the glycocalyx as well as in the goblet cells. Freitas et al. maximus acting as mechanosensors, mediating in neurotransmission, (2002) showed that the lectin histochemical pattern of the secreted facilitating the propagation of electrical events or acting as pacemaker mucins. We suggest that the differences found in the present study cells. Even though the ICC from the submucosal surface of the circular may be also the result of the interaction between mucins and bacteria muscle (ICC-SM) have been described as slow wave generating cells in ’ transported through the colonic groove. other s colon (Mazzone & Farrugia, 2007), our results demon- strated that this subcellular type is within the ascending colon, restricted just to the colonic groove of L. maximus. Since Takahashi & 4.3 | Distribution pattern of the interstitial Sakaguchi (2000) have determined that the groove transports mainly cells of Cajal bacteria in a retrograde way by antiperistaltic mechanisms, it is possible The motor activity of the gastrointestinal tract (GIT) is a complex physi- that the motor activity of this region may be related to the ICC-SM ological process that implies the interaction of three cell types: the singular distribution pattern. 12 | TANO DE LA HOZ ET AL.

the experiments and analyzed data. C.N.Z. provided technical assis- tance with IHC staining and participated in the analysis of the results. All authors wrote the manuscript and approved the final submission.

REFERENCES Accili, D., Menghi, G., & Gabrielli, M. G. (2008). Lectin histochemistry for in situ profiling of rat colon sialoglycoconjugates. Histology and Histo- phatology, 23, 863–875. Angata, T., & Varki, A. (2002). Chemical diversity in the sialic acids and related a-keto acids: An evolutionary perspective. Chemical Reviews, 102, 439–469. Bansil, R., & Turner, B. S. (2006). Mucin structure, aggregation, physio- logical functions and biomedical applications. Current Opinion in Col- loid & Interface Science, 11, 164–170. Beyaz, F., & Liman, N. (2009). The prenatal development and histochem- istry of the ileal mucins in the bovine fetuses. Anatomia, Histologia, Embryologia, 38, 436–442. Bjornhag,€ G., & Snipes, R. L. (1999). Colonic separation mechanism in lagomorph and rodent species—A comparison. Mitteilungen Aus Dem Museum Fur€ Naturkunde in Berlin, 75, 275–281. Bontti, E. E., Boo, R. M., Lindstrom,€ L. I., & Elia, O. R. (1999). Botanical composition of cattle and vizcacha diets in central Argentina. Journal of Range Management, 52, 370–377. Branch, L. C., Villarreal, D., Sbriller, A. P., & Sosa, R. A. (1994). Diet selec- tion of the plains vizcacha (Lagostomus maximus, family ) in relation to resource abundance in semi-arid scrub. Canadian Jour- nal of Zoology, 72, 2210–2216. Clauss, M., Besselmann, D., Schwarm, A., Ortmann, S., & Hatt, J. M. (2007). Demonstrating coprophagy with passage markers? The exam- ple of the plains viscacha (Lagostomus maximus). Comparative Bio- chemistry and Physiology Part A: Molecular & Integrative Physiology, 147(2), 453–459. Commission on Life Sciences National Research Council. (1996). Institute of laboratory animals resources. Guide for the care and use of laboratory animals. Washington, DC: Editorial National Academic Press. Corfield, A. P. (2015). Mucins: A biologically relevant glycan barrier in mucosal protection. Biochimica Et Biophysica Acta, 1850, 236–252. Culling, C. F. A., Reid, P. E., & Dunn, W. L. (1976). A new histochemical method for the identification and visualization of both side-chain acylated and non-acylated sialic acids. Journal of Histochemistry and Cytochemistry, 24, 1225–1230. Forstner, J. F., Oliver, M. G., & Sylvester, F. A. (1995). Production, struc- FIGURE 10 Lagostomus maximus, positive c-kit cells of the submu- cosal border. (a–c) Detail of the ICC-SM. Arrow head, ICC-SM; ture and biologic relevance of gastrointestinal mucins. In M. J. Blaser, P. D. Smith, J. I. Ravdin, H. B. Greenberg, & R. L. Guerrant (Eds.), arrow, cytoplasmic prolongation; SubM, tunica submucosa; TM, Infections of the gastrointestinal tract (pp. 71–88). New York: Raven tunica muscularis. Scale bar: 30 mm Press. ACKNOWLEDGMENTS Freitas, M., Axelsson, L. G., Cayuela, C., Midtvedt, T., & Trugnan, G. (2002). Microbial–host interactions specifically control the glycosyla- This research was supported by a Grant from Universidad Nacional tion pattern in intestinal mouse mucosa. Histochemistry and Cell Biol- de Mar del Plata (EXA 765/16), Buenos Aires, Argentina. M.F.T.d.l.H ogy, 118, 149–161. and C.N.Z are members of CONICET, Argentina. We thank the staff Hagen, K. B., Besselmann, D., Cyrus-Eulenberger, U., Vendl, C., Ortmann, of Estacion de Cría de Animales Silvestres (ECAS), Ministerio de S., Zingg, R., ... Clauss, M. (2015). Digestive physiology of the plains viscacha (Lagostomus maximus): A large herbivorous hystricomorph Agroindustria de la Provincia de Buenos Aires. rodent. Zoo Biology, 34(4), 345–359. Hansson, G. C. (2012). Role of mucus layers in gut infection and inflam- AUTHOR CONTRIBUTIONS mation. Current Opinion in Microbiology, 15,57–62. The study was designed by A.O.D. and A.M.F. and was undertaken Jackson, J. E., Branch, L. C., & Villarreal, D. (1996). Lagostomus maximus. by M.F.T.d.l.H. as part of her doctoral thesis. M.F.T.d.l.H. performed Mammal Species, 543,1–6. TANO DE LA HOZ ET AL. | 13

Kim, Y. S., & Ho, S. B. (2010). Intestinal goblet cells and mucins in health tract. Mechanism and distribution of the reactive substance. Journal and disease: Recent insights and progress. Current Gastroenterology of Histochemistry and Cytochemistry, 21, 473–482. – Reports, 12, 319 330. Roberton, A. M., & Wright, D. P. (1997). Bacterial glycosulphatases and Kotze, S. H., van der Merwe, E. L., Ndou, R., O’riain, M. J., & Bennett, N. sulphomucin degradation. Canadian Journal of Gastroenterology, 11, C. (2009). The colonic groove or furrow: A comparative morphologi- 361–366. cal study of six species of African mole-rats (Rodentia, Bathyergidae). Sakaguchi, E. (2003). Digestive strategies of small hindgut fermenters. – Journal of Morphology, 270, 966 975. Animal Science Journal, 74, 327–337. ’ Kotze, S. H., van der Merwe, E. L., & O riain, M. J. (2006). The topogra- Sanders, K. M., Kito, Y., Hwang, S. J., & Ward, S. M. (2016). Regulation phy and gross anatomy of the gastrointestinal tract of the Cape of gastrointestinal smooth muscle function by interstitial cells. Physi- Dune mole-rat (Bathyergus suillus). Anatomia, Histologia, Embryologia, ology, 31, 316–326. 35, 259–264. Sanders, K. M., Ord€ og,€ T., Koh, S. D., Torihashi, S., & Ward, S. M. (1999). Lev, R. A., & Spicer, S. S. (1964). Specific staining of sulphate groups Development and plasticity of interstitial cells of Cajal. Neurogas- with Alcian Blue at low pH. Journal of Histochemistry and Cytochemis- troenterology & Motility, 11, 311–338. try, 12, 309–309. Scillitani, G., & Mentino, D. (2015). Comparative glycopattern analysis of Liquori, G. E., Mastrodonato, M., Mentino, D., Scillitani, G., Desantis, S., mucins in the Brunner’s glands of the guinea-pig and the house Portincasa, P., & Ferri, D. (2012). In situ characterization of O-linked mouse (Rodentia). Acta Histochemica, 117, 612–623. glycans of Muc2 in mouse colon. Acta Histochemica, 114, 723–732. Snipes, R. L., Hornicke,€ H., Bjornhag,€ G., & Stahl, W. (1988). Regional dif- Lison, L. (1953). Histochimie et cytochimie animales. Principes et methodes ferences in hindgut structure and function in the nutria, Myocastor (607 p.). Paris: Gauthier-Villars. coypus. Cell and Tissue Research, 252, 435–447. Martino, N. S., Zenuto, R. R., & Busch, C. (2007). Nutritional responses to Takahashi, T., & Sakaguchi, E. (1998). Behaviors and nutritional impor- different diet quality in the subterranean rodent Ctenomys talarum tance of coprophagy in captive adult and young nutrias (Myocastor – (tuco-tucos). Comparative Biochemistry and Physiology A, 147, 974 982. coypus). Journal of Comparative Physiology B, 168, 281–288. Mastrodonato, M., Mentino, D., Liquori, G. E., & Ferri, D. (2013). Takahashi, T., & Sakaguchi, E. (2000). Role of the furrow of the proximal Histochemical characterization of the sialic acid residues in mouse colon in the production of soft and hard feces in nutrias, Myocastor – colon mucins. Microscopy Research and Technique, 76, 156 162. coypus. Journal of Comparative Physiology B, 170, 531–535. Mazet, B. (2015). Gastrointestinal motility and its enteric actors in Takahashi, T., & Sakaguchi, E. (2006). Transport of bacteria across and € mechanosensitivity: Past and present. Pflugers Archiv European Journal along the large intestinal lumen of guinea pigs. Journal of Comparative – of Physiology, 467, 191 200. Physiology B, 176, 173–178. Mazzone, A., & Farrugia, G. (2007). Evolving concepts in the cellular con- Tano de la Hoz, M. F., Flamini, M. A., & Díaz, A. O. (2014). Histological trol of gastrointestinal motility: Neurogastroenterology and enteric and histochemical study of the duodenum of the plains viscacha sciences. Gastroenterology Clinics of North America, 36, 499–513. (Lagostomus maximus) at different stages of its ontogenetic develop- Mc Manus, J. F. A. (1948). Histological and histochemical uses of peri- ment. Acta Zoologica, 95,21–31. odic acid. Stain Technology, 23,99–108. Tano de la Hoz, M. F., Flamini, M. A., & Díaz, A. O. (2016). Comparative McGuckin, M. A., Linden, S. K., Sutton, P., & Florin, T. H. (2011). Mucin analysis of the morphology, ultrastructure, and glycosylation pattern dynamics and enteric pathogens. Nature Reviews Microbiology, 9, of the jejunum and ileum of the wild rodent Lagostomus maximus. 265–278. The Anatomical Record, 299, 630–642. Mess, A., & Ade, M. (2005). Feeding biology of the dassie-rat Pteromus Volz, D., Reid, P. E., Park, C. M., Owen, D. A., & Dunn, W. L. (1987). A typicus in captivity. Belgian Journal of Zoology, 135,45–51. new histochemical method for the selective periodate oxidation of – Mowry, R. W. (1963). The special value of methods that colour both total tissue sialic acids. Histochemistry Journal, 19, 311 318. acidic and vicinal hydroxyl groups in the histochemical study of Zuniga,~ M., Tur Marí, J., Milocco, S., & Pineiro, R. (2001). Ciencia y mucins with revised directions for the colloidal iron stain, the use Tecnología en Proteccion y Experimentacion Animal. Madrid: Editorial of Alcian blue 8GX, and their combination with the periodic acid- McGraw-Hill Interamericana. Schiff reaction. Annals of the New York Academy of Sciences, 106, 402–423. Pearse, A. G. E. (1985). Histochemistry: Theoretical and applied (Vol. 2, How to cite this article: Tano de la Hoz MF, Flamini MA, 631 p.). Edinburgh: Churchill Livingstone. Zanuzzi CN, Díaz AO. The colonic groove of the plains viscacha Puig, S., Videla, F., Cona, M., Monge, S., & Roig, V. (1998). Diet of the (Lagostomus maximus): Histochemical evidence of an abrupt vizcacha Lagostomus maximus (Rodentia, Chinchillidae), habitat prefer- change in the glycosylation pattern of goblet cells. Journal of ences and food availability in Northern Patagonia. Argentina. Mam- – malia, 62, 191–204. Morphology. 2017;00:000 000. https://doi.org/10.1002/jmor. Reid, P. E., Culling, C. F. A., & Dunn, W. L. (1973). Saponification induced 20735 increase in the periodic acid Schiff reaction in the gastrointestinal